checkpoint-150/trainer_state.json

{
  "best_global_step": 150,
  "best_metric": 0.3620273470878601,
  "best_model_checkpoint": "/content/output/checkpoint-150",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 1.7940553426742554,
      "learning_rate": 5.917159763313609e-06,
      "loss": 1.3972,
      "mean_token_accuracy": 0.7618847399950027,
      "num_tokens": 31152.0,
      "step": 10
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 1.0511088371276855,
      "learning_rate": 1.1834319526627219e-05,
      "loss": 1.2288,
      "mean_token_accuracy": 0.7673537939786911,
      "num_tokens": 61773.0,
      "step": 20
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.8096975088119507,
      "learning_rate": 1.7751479289940828e-05,
      "loss": 0.9255,
      "mean_token_accuracy": 0.7948055118322372,
      "num_tokens": 94758.0,
      "step": 30
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.6213255524635315,
      "learning_rate": 2.3668639053254438e-05,
      "loss": 0.7366,
      "mean_token_accuracy": 0.8212289035320282,
      "num_tokens": 125915.0,
      "step": 40
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.42887863516807556,
      "learning_rate": 2.958579881656805e-05,
      "loss": 0.5556,
      "mean_token_accuracy": 0.8463607966899872,
      "num_tokens": 159364.0,
      "step": 50
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5108550786972046,
      "learning_rate": 3.5502958579881656e-05,
      "loss": 0.4676,
      "mean_token_accuracy": 0.8614094287157059,
      "num_tokens": 190374.0,
      "step": 60
    },
    {
      "epoch": 0.9333333333333333,
      "grad_norm": 0.4310983121395111,
      "learning_rate": 4.142011834319527e-05,
      "loss": 0.4214,
      "mean_token_accuracy": 0.8710317760705948,
      "num_tokens": 222156.0,
      "step": 70
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.4043419659137726,
      "eval_mean_token_accuracy": 0.8734077858924866,
      "eval_num_tokens": 238011.0,
      "eval_runtime": 50.8925,
      "eval_samples_per_second": 1.965,
      "eval_steps_per_second": 0.491,
      "step": 75
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 0.35767048597335815,
      "learning_rate": 4.7337278106508875e-05,
      "loss": 0.3901,
      "mean_token_accuracy": 0.8787548273801804,
      "num_tokens": 255825.0,
      "step": 80
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.3684692680835724,
      "learning_rate": 5.3254437869822495e-05,
      "loss": 0.3956,
      "mean_token_accuracy": 0.8787799149751663,
      "num_tokens": 287370.0,
      "step": 90
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.35148611664772034,
      "learning_rate": 5.91715976331361e-05,
      "loss": 0.3822,
      "mean_token_accuracy": 0.8814697057008744,
      "num_tokens": 315883.0,
      "step": 100
    },
    {
      "epoch": 1.4666666666666668,
      "grad_norm": 0.39074671268463135,
      "learning_rate": 6.50887573964497e-05,
      "loss": 0.3703,
      "mean_token_accuracy": 0.8836676925420761,
      "num_tokens": 347324.0,
      "step": 110
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.38130292296409607,
      "learning_rate": 7.100591715976331e-05,
      "loss": 0.3762,
      "mean_token_accuracy": 0.8851484537124634,
      "num_tokens": 379374.0,
      "step": 120
    },
    {
      "epoch": 1.7333333333333334,
      "grad_norm": 0.4091486632823944,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.3446,
      "mean_token_accuracy": 0.8894414573907852,
      "num_tokens": 411843.0,
      "step": 130
    },
    {
      "epoch": 1.8666666666666667,
      "grad_norm": 0.361017644405365,
      "learning_rate": 8.284023668639054e-05,
      "loss": 0.3494,
      "mean_token_accuracy": 0.8915452778339386,
      "num_tokens": 444017.0,
      "step": 140
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3487047851085663,
      "learning_rate": 8.875739644970414e-05,
      "loss": 0.3674,
      "mean_token_accuracy": 0.8846240967512131,
      "num_tokens": 476022.0,
      "step": 150
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.3620273470878601,
      "eval_mean_token_accuracy": 0.8875497984886169,
      "eval_num_tokens": 476022.0,
      "eval_runtime": 50.8961,
      "eval_samples_per_second": 1.965,
      "eval_steps_per_second": 0.491,
      "step": 150
    }
  ],
  "logging_steps": 10,
  "max_steps": 1125,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.842489979664384e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}